-
- Downloads
Updated headerinformation to recent standard. There is a small error in the...
Updated headerinformation to recent standard. There is a small error in the header information, namely “The largest pi = 2 * min**2 = 2**(c_dsp_dat_w-1) …” is not the right equation. It should be = 2**(c_dsp_prod_w-1) or = 2 * ((-2)**(c_dsp_dat_w-1))**2, and that is equal to = (-2)**(2*(c_dsp_dat_w-1)) = 2* 4*(c_dsp_dat_w-1). I updated it to 2**(c_dsp_prod_w-1). Added five (extra) libraries with two commented out: --LIBRARY ip_agi027_xxxx_mult_lib; --LIBRARY ip_agi027_xxxx_mult_rtl_lib; library ip_agi027_xxxx_complex_mult_altmult_complex_1910; library ip_agi027_xxxx_complex_mult_rtl_lib; library ip_agi027_xxxx_complex_mult_rtl_canonical_lib; Add generate-block inclusive the instantiation of a module for the agi027_xxxx: -- IP variants for <= 18 bit: ip_agi027_xxxx_complex_mult -- IP variants for > 18 bit and <= 27 bit: ip_agi027_xxxx_complex_mult_27b -- RTL variants that can infer multipliers for a technology, fits all widths: ip_agi027_xxxx_complex_mult_rtl -- RTL variants that can infer multipliers for a technology, fits all widths: ip_agi027_xxxx_complex_mult_rtl_canonical.
parent
405b73a8
No related branches found
No related tags found
Loading
Please register or sign in to comment